A. Field of the Invention PA1 B. Related Art
The present invention relates generally to the fields of molecular biology, gene regulation and pathology. More specifically, in certain aspects, the invention relates to the identification of premalignant or malignant conditions in tissues. In other aspects, the invention relates to methods and compositions for the identification of apoptosis, or programmed death, in cells. In particular, the present invention relates to monitoring of levels of apurinic/apyrimidinic endonucleases, also known as APE's.
(i) Cancer Markers
Despite continued efforts worldwide to identify useful prognostic factors for premalignant and malignant conditions (hereinafter together referred to as "(pre)malignant conditions") of human tissues, relatively few markers and associated screens have been discovered which reliably identify (pre)malignant conditions. As one specific example, squamous cell carcinoma of the cervix uteri (SCCC) is the second most common female malignancy, and the leading cause of death by cancer in women worldwide (Mitchell et al., 1995; Burger et al., 1995; Richard et al., 1995). Based on recent estimates, there will be approximately 15,800 new cases of invasive disease, 65,000 cases of carcinoma in situ (CIS; premalignant), and 4800 deaths attributed to SCCC annually in the United States alone (American Cancer Society, 1995). African- and Hispanic-American women and poor Caucasian women were found to have a mortality rate from cervical cancer of more than double that of all Caucasian-American women (Burger et al., 1993; Miller et al., 1993; Davis et al., 1995; Parham et aL 1995), and HIV positive women were diagnosed with CIS at five times the rate of HIV negative women (Wright et al., 1994; Heard et al., 1995).
Cervical cancer arises in the squamous cells lining the cervix tissue. Precancerous lesions are known as CIS, dysplasia, or cervical intraepithelial neoplasia (CIN). Although the development of these cells into invasive carcinoma can take ten to twelve years, in about 10% of patients the development is much more rapid, occurring in less than a year (National Cancer Institute, 1995). Early detection of cervical cancer substantially increases the probability of survival, with both malignant and premalignant conditions being detectable by the so-called Pap smear.
While the Pap smear is relatively widely used, and has had an overall positive impact on women's health, it presents several significant drawbacks. Pap smear sampling must be performed by highly trained clinicians to result in an interpretable, representative sample of the cells lining the cervix (Koss, 1989). As well, a trained cytologist must analyze the morphology of the cells upon microscopic examination (Koss, 1989). Significant human error is attributed to both steps, contributing to high levels of false-negative readings (Koss, 1989; Koss, 1993). A majority of studies estimate the rate of false-negatives at 20%-30%, with various other studies putting this value at 5% to in excess of 50% (Morell et al., 1982). In addition, subsequent to a positive interpretation of a Pap smear, a physician typically biopsies the cervical tissue to confirm the diagnosis and assist in the determination of the stage of the disease and the design of an appropriate treatment regimen. The biopsy sample is analyzed by a pathologist for the presence or absence of (pre)malignant cells and to determine the extent of tumor growth. Human error can also arise in these procedures (Sideri, et al., 1982).
In light of these and other shortcomings of the common Pap smear, researchers have been actively seeking a reliable marker for (pre)malignant states in cervical tissue. A useful marker and associated assay are understood to require a number of attributes. An assay using the marker must consistently detect differences in cancer and noncancer, and exhibit both specificity (few false positives) and sensitivity (few false negatives). Quantitative assays find increased utility over those which are merely qualitative, and cancer marker specific for a particular organ or cell type will be more useful for initial screening purposes, but organ/cell specificity is less important for monitoring previously diagnosed patients.
A number of putative markers for (pre)malignant conditions of the cervix have been identified; however, the markers suggested to date exhibit several shortcomings. For instance, squamous cell carcinoma antigen is a glycoprotein purified from SCCC that has been found to be a marker for cancerous conditions of the cervix (Kato et al., 1982; Kato et al., 1984). This marker was originally called T-4 in a lesser-purified form, and serum SCCA was found to be elevated in 61% of SCCC cases overall, ranging from 30-45% in Stage 1 to 90-100% in Stage 4 (Crombach et al., 1989) (FIGO classification, National Cancer Institute, 1995). In original testing of SCCC as a tissue marker using flow cytometry of vaginal smear cells, 85% of SCCC cases, 80% of severe and 43% of mild to moderate dysplasias and 21% of normal specimens contained cells stained with antibodies to SCCA (Suehiro et al., 1986). This lack of specificity decreases the usefulness of SCCA as a marker, which has also been bolstered by the observation that cytosolic concentration of SCCA in normal cells is twice as high in normal cells than in SCCC cells (Crombach et al., 1989).
Another putative marker for SCCC is carcinoembryonic antigen (CEA). One of the most studied antigens using immunohistochemical analysis for the determination of neoplastic cells in SCCC is CEA. Reports as to the percentage of different dysplastic and neoplastic lesions stained have varied (Toki et al., 1991; Rutenan et al., 1978; van Nagell et al., 1982; Bamford et al., 1983; Bychkov et al., 1983; McDiken et al., 1983; Lindgren et al, 1986; Agarwal et al., 1990). Additional possible markers which have been studied, with varying degrees of success, include proliferating cell nuclear antigen (PCNA) (Raju, 1994; Steinbeck et al., 1995), epithelial membrane antigen (EMA) (Bamford et al., 1983; Sarker et al., 1994), various keratins (Rajur et al., 1988; Auger et al., 1990), Tn antigen (Hamada et al., 1993; Hirao et al., 1993), oncogenes and tumor suppressor genes (Kohier et al., 1989; Tervahauta et al., 1994; Hale et al., 1993; Terzano et al., 1993; Sainz et al., 1993; Tervahauta et al., 1993; Cardillo et al., 1993), and various others (Fuchs et al., 1989; Flint et al., 1988; Costa et a., 1987; Lara et al., 1994; Carico et al., 1993; Harlozinski et al., 1985).
(ii) APE
Apurinic/apyrimidinic endonucleases (hereinafter sometimes referred to as "apurinic endonuclease" or "APE") catalyze repair of baseless sites in DNA. At least 10,000-20,000 of these sites are generated daily in every human cell as a result of oxidation, spontaneous hydrolysis, and the removal of modified bases by DNA glycosylases (Loeb, 1985, FIG. 15). These baseless sites disrupt transcription and are highly mutagenic if not repaired.
The major human apurinic/apyrinmidinic endonuclease is a 37,000 Dalton protein which has been cloned and shown to complement APE deficient bacteria (Demple et al., 1991). APE has been shown to be identical to Ref-1, a redox factor facilitating the DNA binding of a number of transcription factors, many of which are important in oncogenesis, including Fos, Jun, Myb, and members of the ATF/CREB family (Xanthoudakis et al., 1992). Recently, APE has also been shown to be involved in the negative regulation of transcription of the parathyroid hormone gene by extracellular calcium in vitro (Okazaki et al., 1994). Ape also appears to be a major regulator of p53 activity, acting through protein modification of p53 (Jayaraman et al., 1997)
Besides DNA repair activity, the major human APE repair enzyme has been found to exhibit multiple flnctions, many by in vitro studies. For example, investigators studying Ref-1, a redox regulating transcription factor, discovered that Ref-1 and APE were identical (Xanthoudakis et al., 1992). APE/Ref-1 facilitates the DNA binding characteristics of Jun-Jun homodimers, Fos-Jun heterodimers, HeLa AP-1, and numerous other transcription factors, including Myb, members of the CREB family and nuclear factor-.kappa.B (Xanthoudakis et al., 1992).
Immunohistochemistry has been used to examine the subcellular distribution of APE in several different human tissues. The results show that levels vary significantly in different tissues (Duguid et al., 1995). APE expression in skin and intestine was tightly linked to cellular maturation. In most tissues, APE was detected primarily in the nucleus, where the APE staining pattern followed that of chromatin. In hepatocytes and some neurons, however, APE was detected primarily in the cytoplasm.
At this point in time, APE has not been associated with any particular pathologic conditions. Though clearly important to cellular function, specific diseases resulting from aberrations in this protein's function are not known.